By Andrea Hak Arguably the most important thing that the rise of intelligent AI could potentially bring is access. Access to goods, services, and information not just for the few, but for everyone. Victoria Slivkoff, Head of Ecosystem at Walden Catalyst and Managing Director of Extreme Tech Challenge — a nonprofit uniting startups and VCs to accelerate progress toward the UN Sustainable Development Goals (SDGs) — is excited for what lies ahead. In her view, the physical manifestation of AI could bring us closer to realising these ambitious goals. “Now we’re moving into the area of reasoning. AI is not just aggregating and…This story continues at The Next Web

Source:: The Next Web

By Siôn Geschwindt “Microbial protein,” says Katelijne Bekers, waving a vial of beige-coloured powder in front of me like it’s a magic potion. It doesn’t look like your typical lunch fare, but this unassuming dust could play a crucial role in the future of food. Bekers is the co-founder of MicroHarvest, a Hamburg and Lisbon-based startup that turns agricultural waste streams into protein powder using microbes — tiny organisms that exist all around us. The vegan ingredient is already making its way into dog treats. If all goes to plan, human snacks like protein bars, shakes, and ice cream won’t be far behind.…This story continues at The Next Web

Source:: The Next Web

Apple Intelligence will lean heavily into Apple’s existing work with App Intents and Shortcuts integrations. The app services App Intents provide will be combined with Apple Intelligence and supported by what your device knows about where you are, what you need and the things you usually do.

(It will all take place privately and on device, of course.)

What are App Intents?

“The App Intents framework provides functionality to deeply integrate your app’s actions and content with system experiences across platforms, including Siri, Spotlight, widgets, controls and more,” Apple has explained. That means App Intents can integrate your apps actions with Siri, Spotlight, and other apps, making it easier for users to get things done.

First announced in 2024, App Intents are already available across every app — and while not every app supports App Intents, Apple really wants developers to climb aboard. 

During a WWDC 2025 session, the company suggested App Intents as a way to make the key functionalities of your app available across the system. It calls those functionalities the “verbs” of any app. 

These can be combined with those from other third-parties and Apple apps for useful tasks, allowing users to access functionalities available in other apps from within a developer’s own app. That means offering users customized Spotlight results, custom actions for Apple Pencil Pro, contextually aware commands for the Action Button, interactive widgets and more. 

What should be interesting is that App Intents will make it possible to execute complex strings of actions/verbs on demand; for developers, App Intents is a two-way street enabling them to build far more complex app experiences than they could do alone, while also giving an app the power to reach out to users via other apps.

What about Siri?

App Intents are both Shortcuts and Siri compatible. And that’s really where I see them begin to shine, as it means App Intents can be executed via voice commands. Apple watcher Mark Gurman goes as far as to say that once complete, they will permit you to fully control your iPhone using only voice, including finding, editing, and sharing a photo.

In other words, you’ll be able to string Intents/Shortcuts from across multiple apps together on your behalf so you can get more complex tasks done just by asking your device. 

This was certainly what Apple’s 2024 announcement of a smarter Siri promised. And while that work has been seriously delayed, the company recently said it is going well, and it hopes to introduce the new tools in spring.

Unlocking the apps

The sticking point for App Intents is that they require app functionality (those “verbs”) to be unlocked and made available using Apple’s own APIs. The problem there is that some developers might be resistant to making such functionality available outside of their own app, as they fear loss of user engagement.

I think that resistance is part of the reason Apple is working with the big name developers behind some of the world’s most widely used apps. Bloomberg’s senior Apple sleuth tells us it is working with Uber, AllTrails, Threads, Amazon, Temu, YouTube, Facebook, and WhatsApp on this, so once these new features do appear, the “verbs” for the most widely used apps will be supported by the system. 

Of course, the danger here is that as the big apps become even more omnipresent across Apple’s and other systems, the opportunities for smaller third-party apps to intrude into the experience they provide will erode. After all, how do you make any task more convenient than asking Siri to make it happen? Will we end up with an iOS ecosystem that’s as commoditized as the web seems to have become, with only a few brands occupying the majority of online attention? What impact would such an outcome have on digital economies, particularly as online attention is detected and solutions provided almost automatically through AI?

Existential queries aside, working with these key developers will also give Apple better insight into any flaws in its software that might need rectifying as it moves toward the first public beta of these new Apple Intelligence features working through Siri. Given the challenges it has faced getting to this point, working with others might help its teams deliver on time. 

I’m interested to see how this unfolds.

You can follow me on social media! Join me on BlueSky,  LinkedIn, and Mastodon.

Source:: Computer World

Nvidia has developed a generative AI (genAI) model to help robots make human-like decisions by analyzing surrounding scenes.

The Cosmos Reason model in robots can take in information from video and graphics input, analyze the data, and use its understanding to make decisions.

Cosmos Reason, announced on Monday,  helps robots “think like humans do” and make decisions with “just common sense,” said Rev Lebaredian, vice president of Omniverse and simulation technologies.

The model is lightweight at 7 billion parameters and can be used in a variety of physical devices such as installed cameras, traffic signals, and instruments in factories.

“Every smart IoT device that can see, from cameras to traffic lights, every home or industrial robot, will have reasoning,” Lebaredian said.

Companies can develop video AI agents, which will act on massive amounts of data gathered and analyzed from recorded video data and livestreams. “These video agents will soon be everywhere, automating traffic monitoring, improving safety, and enhancing video inspection in everything from industrial facilities to entire cities,” Lebaredian said.

Cosmos Reason is what Nvidia calls a “vision language model” (VLM). That’s different from typical text-based models, which can generate images, videos, or text.

Nvidia’s Cosmos Reason VLM is designed to help robots make better decisions.
Nvidia

OpenAI and other companies have released VLMs, but Cosmos Reason can do deeper reasoning on a long tail of unseen scenarios, he said. The models can establish prior understanding of scenarios and take into account physical interactions and then infer complex interactions or motivations of objects and actors in the scene. It can also understand new and unseen experiences.

For example, robots will be able to connect the dots of making toast, understanding that toast requires butter and a toaster — and a plate on which to serve the food.

Today’s AI robot models have two types of technology underpinning their activity. The VLM interprets instructions and plans actions, while “vision language action” allows for fast actions and muscle memory.

Cosmos Reason is open-source and now available for download, the company said, but it will only work on Nvidia’s hardware. 

The company sells the Jetson Thor DGX computer for robots and said its new RTX Pro 6000 GPUs will be in high-end servers. The company also announced new RTX Pro 4000 and 2000 GPUs for high-end desktops. The new GPUs are based on the Blackwell architecture.

Nvidia is grouping its world-building and simulation products under the Omniverse product line. Cosmos Reason is one of many models developed by the company to improve productivity in factories, warehouses, robots, vehicles, and other physical locations.

Omniverse products involve creating a digital copy representation of physical products in the real world. Information in the virtual world is used to create synthetic data to train vision language models.

Source:: Computer World

By Deepti Pathak OPPO has officially launched the K13 Turbo Pro 5G and K13 Turbo 5G, the first smartphones…
The post First Impressions of the OPPO K13 Turbo appeared first on Fossbytes.

Source:: Fossbytes

By Wouter Moolenaar Crypto crashes, money laundering, and digital fraud — the EU’s financial watchdogs have had enough. Regulatory bodies need to keep up by rolling out tighter regulations aimed at strengthening consumer protections and stabilising the market.  As EU lawmakers scramble to protect consumers, others worry they are smothering growth. Case in point: in 2024, the FCA fined HSBC £6.2mn for not properly treating customers in financial difficulty. The regulatory bodies are defending the public, but had restrictions been lighter, would HSBC have had more creative solutions for its customers, such as embedding personalised, data-first lending? Banks have been fearful of exploring…This story continues at The Next WebOr just read more coverage about: Fintech

Source:: The Next Web

By Hisan Kidwai Free Fire Max is one of the most popular games on the planet, and for good…
The post Garena Free Fire Max Redeem Codes for August 11 appeared first on Fossbytes.

Source:: Fossbytes

By Hisan Kidwai Free Fire Max is one of the most popular games on the planet, and for good…
The post Garena Free Fire Max Redeem Codes for August 9 appeared first on Fossbytes.

Source:: Fossbytes

Apple is going with the wind, allegedly confirming plans to integrate OpenAI’s latest GPT-5 model for ChatGPT within its 26 series of operating system releases for iPhones, iPads, and Macs. Apple Intelligence already integrates with ChatGPT, but only the 4o model. Apple allegedly told 9to5Mac that it intends to integrate GPT-5 when it ships its new operating systems, potentially next month.

OpenAI introduced GPT-5 in early August. “Our smartest, fastest, and most useful model yet, with thinking built in,” the company said.

GPT-5 a ‘significant leap,’ says OpenAI

The ChatGPT maker claims its new AI system delivers a “significant leap” above previous models, featuring a deeper reasoning model for harder problems and the capacity to filter enquiries to the relevant parts of the engine. GPT-5 outperforms previous models on benchmarks and answers questions more swiftly than before, the company also said.

It also hallucinates less, the company said, and “minimizes sycophancy.” (How reassuring.) It seems particularly strong on coding skills, health, and assisting you in creative writing — with or without an em-dash! You can read more about how OpenAI thinks GPT-5 is great in the company press release. The model is available to Plus, Pro, Team, and Free users, with different usage models depending on what you pay.

Apple, of course, provides integration with ChatGPT via Apple Intelligence, which sends requests to OpenAI’s system when Apple can’t handle them alone. Within this integration, users are warned if they are about to use the third-party AI service. Apple has also built in privacy protection so if you use ChatGPT through Apple Intelligence, OpenAI shouldn’t store requests and your IP address should be obscured. If you pay for ChatGPT access, then the service’s own privacy promises apply.

Apple continues to develop its own AI

While Apple has extended support to the new ChatGPT model, it continues to invest in developing its own more targeted AI solutions.

The upcoming operating systems will also introduce Live Translation and Visual Intelligence improvements, and the company has consistently said the contextual intelligence features it promised in 2024 will arrive eventually. Apple CEO Tim Cook recently confirmed the the plan is to introduce these features next year, saying the team is “making good progress on a more personalized Siri.”

Cook’s company is also working to introduce AI solutions for specific tasks. It allegedly intends to introduce a generative AI-powered support assistant tool within the Apple Support app, and may also be preparing to exploit the vast amount of data gathered by the Apple bot for use in a new Answers engine. The latter is described as a “stripped-down” version of ChatGPT, capable of crawling the web to answer questions, and is likely to be used to support Siri, Spotlight, and Safari, as well as being a standalone app. 

“We’ve rarely been first…”

One thing we do know is that Apple has no intention of giving up in the AI race. During a recent meeting, Cook stressed that Apple’s approach isn’t about being first but being best, and that the battle for dominance in the AI space isn’t over yet. Apple has time and power in the race.

“We’ve rarely been first,” he said. “There was a PC before the Mac; there was a smartphone before the iPhone; there were many tablets before the iPad; there was an MP3 player before iPod. But Apple invented the modern versions of those product categories. This is how I feel about AI.”

Cook also promised to make investments to get to where the company wants to get to. We’ve heard plenty of speculation concerning potential takeover targets. While big targets such as Perplexity are frequently mentioned, smaller AI firms such as Runway AI, Eleven Labs, and Pika AI may also prove attractive.

“We’re very open to M&A that accelerates our road map,” Cook said during Apple’s July 25 earnings call. “We basically ask ourselves whether a company can help us accelerate a road map. If they do, then we’re interested.” 

The flipside is that in the event Apple does successfully acquire a company and its talent, it may also need to figure out how to keep them, given the aggressive attempts competitors are making to poach its people.

You can follow me on social media! Join me on BlueSky,  LinkedIn, and Mastodon.

Source:: Computer World

The ambitious $500 billion Stargate AI infrastructure project is moving significantly slower than anticipated, with SoftBank Group CFO Yoshimitsu Goto publicly acknowledging the delays during the company’s Q1 2025 earnings call.

“It’s taking a little longer than our initial timeline,” Goto said during the call, describing the project as proceeding “slower than usual.”

Source:: Computer World

By Deepti Pathak Panasonic has expanded its TV line-up with the launch of the new P-Series range, featuring the…
The post Panasonic Launches Shinobi Pro Mini LED 4K TV Series in India appeared first on Fossbytes.

Source:: Fossbytes

By Deepti Pathak Just started playing Verse Piece and not sure where to find updates or connect with others?…
The post Verse Piece Trello & Discord Link (2025) appeared first on Fossbytes.

Source:: Fossbytes

By Brandon Andrews You’ve spent a lifetime building skills, learning lessons, nurturing relationships, and developing a perspective as prescient and powerful as your personal drive. You’ve poured it all into your business. Now, you have five minutes (or less) to communicate an irresistible vision for the world and convince a panel of respected — and sometimes disrespectful — judges that you can make the vision real and make some money. How do you do it? A pitch competition is a unique moment: I have pitched in, judged, and hosted pitch competitions from Miami to Mongolia. I’m an entrepreneur and investor, and I’ve spent…This story continues at The Next Web

Source:: The Next Web

More than two years after GPT-4’s release, OpenAI has unveiled GPT-5, boasting sharper reasoning, multimodal input, better math skills, and cleaner task execution, according to the company.

The large language model (LLM) — now rolling out to ChatGPT users and available in the API — is “smarter, more stable, and more versatile” and built to handle real-world tasks more like a human expert, OpenAI said.

In anticipation of OpenAI’s new AI model, Anthropic released the latest version of its own chatbot, Claude, earlier in the week.

Claude Opus 4.1 came with improvements particularly in two key areas: it significantly improved its coding capabilities, solving up to 75% of real-world programming tasks based on SWE Verified benchmarks; and the model is capable in detailed research and analysis, especially in tasks that require tracking lots of information and intelligently finding answers, according to Anthropic.

For developers, OpenAI claims GPT-5 is its most powerful coding model to date, outperforming GPT-o3 in benchmarks and real-world tasks. The model is “fine-tuned for agentic tools” like Cursor, Windsurf, Copilot, and Codex CLI, and it set new records in testing, the company stated in a blog.

According to OpenAI, GPT-5 delivers sharper reasoning, handling complex problems and multi-step instructions with greater accuracy and focus. It stays on track, follows directions more precisely, and produces more useful, reliable output, the company said.

Users can also expect to see fewer hallucinations and will have better customization tools, making GPT-5 more dependable and easier to adapt to specific industries and needs, OpenAI said.

It also builds on GPT-4o’s multimodal abilities, offering smoother interactions across text, images, and audio, according to OpenAI.

GPT-5 will be OpenAI’s “most significant do or die moment yet,” according to Nathaniel Whittemore, CEO of Superintelligent, a New York-based AI education platform.

“Ever since the launch of ChatGPT, they’ve been the model state of the art. While competitors like Google and Meta can take advantage of hundreds of millions of existing users to put AI products in front of, OpenAI relies on winning new users by being far ahead of the other AI labs,” Whittemore said.

OpenAI chief operating officer Brad Lightcap said ChatGPT is now in use by more than five million business users — up from three million in June.

Biopharmaceutical company Amgen is one of the early adopters of GPT-5. Sean Bruich, senior vice president of AI & Data at Amgen, said AI only works in science if it meets the highest bar, and “GPT-5 clears it,” delivering sharper accuracy, better context, and faster results across Amgen’s workflows.

“GPT-5… is doing a better job navigating ambiguity where context matters. We are seeing promising early results from deploying GPT-5 across workflows,” he said. He also said the model was faster, more reliable, and had higher quality outputs than GPT-4 and other earlier models.

Ethan Mollick, an associate professor at The Wharton School, had early access to GPT-5. “It is a big deal,” he said in a blog post. He asked the model to do something dramatic to prove that point. The model thought for 24 seconds and then delivered a poetic manifesto of AI capability — specifically, a rhetorical, alliterative showcase of “a multifunctional intelligence system.”

“GPT-5 just does stuff, often extraordinary stuff, sometimes weird stuff, sometimes very AI stuff, on its own. And that is what makes it so interesting,” Mollick said.

After “many AI conversations,” Mollick said he has found two big issues that limit most people’s success in using AI models: First, most people don’t know which model to use — so they get fast, weak results instead of more complete answers from the powerful reasoning models.

“The longer [the models] think, the better the answer, but thinking costs money and takes time. So OpenAI previously made the default ChatGPT use fast, dumb models, hiding the good stuff from most users,” Mollick said. “A surprising number of people have never seen what AI can actually do because they’re stuck on GPT-4o, and don’t know which of the confusingly named models are better.”

Second, most people also don’t know what AI can do or how to ask — especially with newer agentic AIs. GPT-5 fixes both problems by choosing models well and suggesting actions, he said. “It is very proactive, always suggesting things to do.”

GPT-5 is beginning to roll out to ChatGPT Plus, Pro, Team, and Free users, with access for Enterprise and Edu customers coming next week. “Once free users reach their GPT‑5 usage limits, they will transition to GPT‑5 mini,” OpenAI said.

Source:: Computer World

Administrators with hybrid Exchange Server environments are urged by Microsoft and the US Cybersecurity and Infrastructure Security Agency (CISA) to quickly plug a high-severity vulnerability or risk system compromise.

Hybrid Exchange deployments offer organizations the ability to extend the user features and admin controls of the on-prem version of Exchange within Microsoft 365. Hybrid deployment can serve as an intermediate step to moving completely to an Exchange Online organization, Microsoft said.

The benefits include secure mail routing between on-premises and Exchange Online organizations, mail routing with a shared domain namespace (for example, both on-premises and Exchange Online organizations use the @contoso.com SMTP domain) and calendar sharing between on-premises and Exchange Online organizations.

Source:: Computer World

By Thomas Macaulay It’s harvest day at the Karolinska Institute in Stockholm. As sunshine bathes the leafy university campus, scientists inside the labs work under cool fluorescent light. Clad in green protective gear, they tend meticulously to test tubes within hermetically sealed cleanrooms. The containers hold the fruits of today’s labour: mesenchymal stem cells (MSCs). Each cell is barely a quarter the width of a human hair but wields remarkable power. MSCs reduce inflammation, repair damaged tissue, and modulate the immune system. They can treat chronic diseases and delay ageing. They may even prevent illness before it begins. But to become a mainstay…This story continues at The Next Web

Source:: The Next Web

By Hisan Kidwai Free Fire Max is one of the most popular games on the planet, and for good…
The post Garena Free Fire Max Redeem Codes for August 7 appeared first on Fossbytes.

Source:: Fossbytes

Once upon a time there was an amazing little voice assistant that ended up being exclusively available in Apple products. It was called Siri and it was ahead of its time.

Because Siri seemed pretty magical when it first hit the iPhone; it would answer requests, find out information, and even do useful things such as taking photographs or naming songs you heard on the radio.

Available in numerous languages and with a range of male and female speaking voices, Siri remains the most widely distributed on-device chatbot in terms of language support. The assistant also scaled well, eventually appearing across Apple’s product lines. But critics and competitors now agree, Siri was too ambitious and failed to keep up with the times.

Back to the future

Cast your mind back to 2010 when Siri first appeared for iPhone. At that time, research into artificial intelligence (AI) and autonomous technologies, ongoing since the launch of the Stanford Artificial Intelligence Lab (SAIL) in the 60’s, really accelerated. Apple led the charge, at least in terms of media profile, and Siri (which the company acquired soon after its introduction) was a leading-edge challenger in the nascent field

What was great about Siri was its fluffy, friendly image. 

Being an Apple product gave the solution access to a huge market of engaged and happy consumers willing to overcome their general concern at the dystopian application of AI to give the friendly little assistant a try. 

Building acceptance one error at a time

Arguably, Apple’s little assistant helped drive acceptance of technologies that have become critical to today’s cutting edge generative AI (genAI) systems, including:

Speech recognition

The idea of intelligent machines

Devices equipped to listen for your commands 24/7

Fast and real-time access to information on spoken request

Andr even real-time transcription.

Siri’s well-publicized errors actually helped build acceptance. After all, if you think about it, the fact that Siri sometimes made mistakes somehow helped humanize it. It is better to think that AI is stupid than to see it as threateningly smart.

This helped a skeptical public come to terms with AI, even while Apple’s assistant embodied a range of concepts people resisted. The logic was that it couldn’t be too bad if machines had this kind of intelligence built inside, right? It’s not as if they are smart enough to fully understand. Did it really matter if the tech listened to you when you thought it was switched off? 

What use would the information picked up be? (The answer: around 81% of UK consumers now claim to have experienced targeted consumer advertising generated by AI.)

Trust in me

We’ve had many debates on these topics since then — debates that show Apple’s commitment to user privacy in AI to be unique, and under attack from competitors and authoritarians alike. It’s almost as if, when some leaders heard Apple CEO Tim Cook warn this is surveillance, they chose to exploit it as an opportunity, rather than protect against it.

All the same, Siri helped people become more capable of placing trust in AI. 

Years later, OpenAI was introduced to a public already more accepting of such tech. That acceptance was to some extent built on the back of Siri adoption, and the appearance of other big name search assistants across the industry.

That acceptance means around 77% of devices in use today have some form of AI, and roughly 90% of organizations are using AI. Investment in the sector is booming, with the tech giants (Apple, Amazon, Google, Microsoft and Meta) spending $92.17 billion on capital expenditures in Q2 2025 alone. That’s up a whopping 66.67% on the year ago quarter, mainly on the strength of big investments in data centers, servers, and AI infrastructure.

Hope, hype, and history

The hype around AI is growing as fast as the investments.

Firms in the space are signing massive multi-billion dollar deals, governments are investing vast quantities of cash and resources to support AI industry development, and consumers are preparing to pay for all this investment come the inevitable industry collapse. 

Consumers will pay? Just look at history. We know this because that’s what happened following the dotcom boom, the South Sea Bubble collapse, and the financial crisis, when unsustainable investments came before a fall. We know this because by the time the highly probable AI industry collapse happens, the tech will be so deeply intertwined in our daily lives we will be told these companies are strategically important, making them “too big to fail.”

So we will bail them out.

What follows Siri? 

I come to bury Siri, not to praise it, because competitors say it was ambitious and failed to keep up with them. But if you’d never experienced Siri, would you have trusted ChatGPT? Perhaps a little, but not as much.

For the future of Siri, ask whether privacy continue to be baked in, or will governments get their way when it comes to data encryption, in which case no one will be private unless they can afford to be.

If governments do get their way and Apple is made to take privacy out of its algorithms, just how much of a threat will Siri become to other AI services, which already seem to respect privacy less? Siri doesn’t seem to know the answer (yet).

Though it probably has quite a lot of data to help it work one out.

You can follow me on social media! Join me on BlueSky,  LinkedIn, and Mastodon.

Source:: Computer World

OpenAI has released its first open-weight language models since GPT-2, marking a significant strategic shift as the company seeks to expand enterprise adoption through more flexible deployment options and reduced operational costs.

The two new models — gpt-oss-120b and gpt-oss-20b — deliver what OpenAI describes as competitive performance while running efficiently on consumer-grade hardware. The larger model reportedly achieves near-parity with OpenAI’s o4-mini on reasoning benchmarks while running on a single 80 GB GPU, while the smaller variant matches o3-mini performance and can operate on edge devices with just 16 GB of memory.

“This is a bold go-to-market move by OpenAI and is now really open,” said Neil Shah, VP for research and partner at Counterpoint Research. “This move nicely challenges rivals such as Meta, DeepSeek, and other proprietary vendors both for cloud and more specifically edge.”

Open-weight models provide access to the trained model parameters, allowing organizations to run and customize the AI locally, but differ from traditional open-source software by not necessarily including the original training code or datasets.

Architecture designed for enterprise efficiency

The models leverage a mixture-of-experts (MoE) architecture to optimize computational efficiency. The gpt-oss-120b activates 5.1 billion parameters per token from its 117 billion total parameters, while gpt-oss-20b activates 3.6 billion from its 21 billion parameter base. Both support 128,000-token context windows and are released under the Apache 2.0 license, enabling unrestricted commercial use and customization.

The models are available for download on Hugging Face and come natively quantized in MXFP4 format, according to the statement. The company has partnered with deployment platforms, including Azure, AWS, Hugging Face, vLLM, Ollama, Fireworks, Together AI, Databricks, and Vercel to ensure broad accessibility.

For enterprise IT teams, this architecture could translate to more predictable resource requirements and potentially significant cost savings compared to proprietary model deployments. According to the statement, the models include instruction following, web search integration, Python code execution, and reasoning capabilities that can be adjusted based on task complexity.

“This will accelerate adoption of OpenAI models for research as well as commercial use under Apache 2.0 license,” Shah noted, highlighting the strategic value of the licensing approach.

Total cost calculations favor high-volume users

The economics of open-weight deployment versus AI-as-a-service present complex calculations for enterprise decision-makers. Organizations face initial infrastructure investments and ongoing operational costs for self-hosting, but can eliminate per-token API fees that accumulate with high-volume usage.

“The TCO calculation will break even for enterprises with high-volume usage or mission-critical needs where the per-token savings of self-hosting and open weights will eventually outweigh the high initial and operational costs,” Shah explained. “For low usage, AI-as-a-Service will benefit better.”

Early enterprise partners, including AI Sweden, Orange, and Snowflake, have begun testing real-world applications, from on-premises hosting for data security to fine-tuning on specialized datasets, the statement added. The timing aligns with enterprise technology spending expected to reach $4.9 trillion in 2025, with AI investments driving much of that growth.

OpenAI said that it subjected the models to comprehensive safety training and evaluations, including testing an adversarially fine-tuned version of gpt-oss-120b under the company’s Preparedness Framework. Also, its methodology was reviewed by external experts, addressing enterprise concerns about open-source AI deployments.

According to OpenAI’s benchmarks, the models showed competitive performance: gpt-oss-120b achieved 79.8% Pass@1 on AIME 2024 and 97.3% on MATH-500, while demonstrating coding capabilities with a 2,029 Elo rating on Codeforces. The company reported that both models performed well on tool use and few-shot function calling — capabilities relevant for business automation.

Strategic decoupling from Microsoft

The release has significant implications for OpenAI’s relationship with Microsoft, its primary investor and cloud partner. Despite the open-weight approach, Microsoft is bringing GPU-optimized versions of the gpt-oss-20b model to Windows devices through ONNX Runtime, supporting local inference via Foundry Local and the AI Toolkit for VS Code, the statement added.

Shah noted that “OpenAI with this move smartly decouples itself from Microsoft Azure and developers can now attach the open-weights models they have been working on and host it if they want to in the future on other rival clouds such as AWS or Google or even OpenAI-Oracle cloud.”

This strategic flexibility could pressure Microsoft to diversify beyond OpenAI partnerships while providing enterprises with greater vendor negotiating power. “This also now offers higher bargaining power for the enterprise against other AI vendors and even AI-as-a-Service models,” Shah observed.

Enterprise deployment considerations

The shift represents OpenAI’s recognition that enterprise AI adoption increasingly requires deployment flexibility. Organizations in regulated industries particularly value data sovereignty options, while others seek to escape vendor lock-in concerns associated with cloud-dependent AI services.

However, enterprises must weigh operational complexity against cost savings. While hardware requirements may be more accessible than previous generations, organizations need expertise in model deployment, fine-tuning, and ongoing maintenance—capabilities that vary significantly across enterprises.

The company is working with hardware providers, including Nvidia, AMD, Cerebras, and Groq, to ensure optimized performance across different systems, potentially easing deployment concerns for enterprise IT teams.

For IT decision-makers, the release expands strategic options in AI deployment models and vendor relationships. The Apache 2.0 licensing removes traditional barriers to customization while enabling organizations to develop proprietary AI applications without ongoing licensing fees.

“In the end it’s a win for enterprises,” Shah concluded, summarizing the broader market impact of OpenAI’s strategic pivot toward openness in the increasingly competitive enterprise AI landscape.

Source:: Computer World

By Deepti Pathak If you have ever played Roblox games based on anime, you might be familiar with A…
The post AOPG Trello & Discord Link (2025) appeared first on Fossbytes.

Source:: Fossbytes